From bizzaro at bc.edu Mon Sep 6 06:07:46 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:12 2006 Subject: [pyorbit] description Message-ID: <37D39272.3AA88A78@bc.edu> Hey Justin! As you can see, I set up the new mailing list. I want to make announcements on some mailing lists and newsgroups, and so we should have a short description of the pyorbit project: background and goal. Could you come up with a paragraph for me? I'd like to post the announcements here: ORBit maillist Python newsgroup CORBA newsgroup gnome-announce maillist PyGTK (gnome-python) maillist Loci & TOL maillists (of course) Any others? :-) Jeff -- +--------------------------------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Bioinformatics Research, Development & Resources | | | | http://theopenlab.uml.edu/ | +--------------------------------------------------+ From bizzaro at bc.edu Mon Sep 6 06:27:20 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:12 2006 Subject: [pyorbit] web page Message-ID: <37D39708.E3833061@bc.edu> This is the Web page for PyORBit: http://theopenlab.uml.edu/pyorbit/ Justin, you have ownership under the group 'web'. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From bizzaro at bc.edu Tue Sep 7 08:06:44 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:12 2006 Subject: [pyorbit] PyORBit (was: CORBA documentation) References: <37D2B65A.6E7DA057@bc.edu> <37D4B4E8.681EB121@redpoll.pharmacy.ualberta.ca> Message-ID: <37D4FFD4.99DAF9D2@bc.edu> Gary Van Domselaar wrote: > > ORBithon sounds cooler than pyorbit, but sounds more like a dance-music > marathon than python-orbit bindings. Exactly. Or the name of a theater that would hold dance performances :-) > I believe LGPL now stands for the GNU Lesser Public License. From > http://www.gnu.org/copyleft/lesser.html > > > [This is the first released version of the Lesser GPL. It also counts > as the successor of the GNU Library Public License, version 2, hence > the version number 2.1.] > That's right. The name change is for 2 reasons: (1) Richard Stallman wants developers to migrate from the LGPL to the regular GPL, and so the word 'lesser' is used in a derogatory way to discourage use (I'm serious; I did some reading about this). The license even has a clause to allow license change ONLY TO THE GPL. (2) "It's not just for libraries anymore." The LGPL serves an attractive purpose for many developers: Binary-only software can link to LGPL-licensed software by the terms of the LGPL. This is not the case with the GPL. But Stallman would rather not encourage binary-only software in any way, and so he is backstepping on the LGPL. However, if commercial acceptance is important (as with large libraries like GTK+ and GNOME, which are LGPL'd), then so is the availability of the 'lesser' version of the GPL. Otherwise, IMO, developers might go with a BSD-type license for libraries. Loci is LGPL'd. Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From bizzaro at bc.edu Tue Sep 7 15:53:30 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:12 2006 Subject: [pyorbit] [Fwd: [pyorbit] description] Message-ID: <37D56D3A.EFBDA3F0@bc.edu> -------------- next part -------------- An embedded message was scrubbed... From: Justin Bradford Subject: Re: [pyorbit] description Date: Tue, 7 Sep 1999 14:17:16 -0500 (CDT) Size: 2084 Url: http://bioinformatics.org/pipermail/pyorbit/attachments/19990907/84e53152/attachment.mht From bizzaro at bc.edu Tue Sep 7 15:53:51 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:12 2006 Subject: [pyorbit] description References: Message-ID: <37D56D4F.D15F298@bc.edu> Justin Bradford wrote: > > How would you feel about holding off until I have something going? Okay by me. > Also, are there any issues for setting up a CVS account for pyORBit? The way CVS is set up now, anyone with a shell account at theopenlab can make a new module. But one must have an account: there is no anonymous access, which we may want to change in a little while. Also, the new module directory needs to be set to the ownership of group 'cvs', and permissions need to be set to group read/write. For example, if you made a new module, the ownership would be justin.justin. We would have to change that to justin.cvs so that others in the cvs group can contribute. As soon as you make the module, let me know an I'll make the changes. FYI, I copied some CVS instructions below. Item (9) talks about making a new module. (1) Set the environment variable. is your shell account user name. For bash, use $ export CVSROOT=':pserver:@theopenlab.uml.edu:/home/cvs' For csh, use $ setenv CVSROOT ':pserver:@theopenlab.uml.edu:/home/cvs' (2) Login. $ cvs login CVS will ask for your password. This will be your shell account pw. (3) To "checkout" a module, use the command by the same name. $ cvs checkout This will create a directory called ./ (within the pwd) with everything in it. (4) You can cd to the new directory and make whatever changes you want. (5) When you add or remove a file, you need to notify CVS. Let's say you created a README file. $ cvs add README (6) Once you have finished working in the directory, commit the changes. $ cvs commit This will commit everything in the pwd, since no specific file was specified. You can also specify a file. (7) Newer versions of CVS support logouts. $ cvs logout (8) You can start all over again at a later time, but it is likely the module will have been changed by someone else. In the directory where you created the directory, rename to something else (say "-old") if you want to keep it. Otherwise typing "cvs checkout " may change some things. If you want to update the copies of the files you have, you can use the "update" command. $ cvs update Remember that the checkout command is used in the parent directory to the directory. The other commands are used within the directory itself. (9) You can also make a new module with the "import" command. $ cvs import -m "" Note that the -m option lets you put in a message without starting an editor. Otherwise CVS will start up $EDITOR. (10) If you would like to see a list of the cvs commands, you can use the "help" command. $ cvs help Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From justin at ukans.edu Wed Sep 8 02:43:45 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:12 2006 Subject: [pyorbit] other binding project Message-ID: Somebody on comp.lang.python sent me this excerpt from the fnorb list: Return-Path: Date: Fri, 7 May 1999 21:46:47 +0800 (CST) From: Michael Robinson Subject: Re: From Fnorb to ORBit... >I could be interrested into another Python ORB with more speed (our project >is for data acquisition), have you an URL about your work (seen nothing on >the ORBit page, neither on the netrinsics.com pages). The Python ORBit binding is under heavy development. The IDL compiler is almost finished for very simple interfaces (basic function call/return of simple types, no structs, no unions, no out/inout arguments). I plan to have something usable within a week or two. I'll make an announcement on the Fnorb and ORBit lists when it gets to that state. If you can provide a minimal list of CORBA features you require for your application, that will help me to prioritize development. -Michael Robinson The date is early May, and I've never heard anything about this guy's bindings, so I imagine the project stalled. Regradless, I'm going to email him and see what he's up to... Justin From justin at ukans.edu Wed Sep 8 04:12:07 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:12 2006 Subject: [pyorbit] other binding project In-Reply-To: Message-ID: > The date is early May, and I've never heard anything about this guy's > bindings, so I imagine the project stalled. Regradless, I'm going to email > him and see what he's up to... I got a response back (as well as his source). I'm going to probably stick what he has in CVS tomorrow (assuming he agrees to the LGPL), and then we can go from there. His code may need some modification to conform with the Python binding proposal, but I doubt it will be too major. This is a major jumpstart to this little project, as we'll be starting with bindings that actually work (or are very near) for simple interfaces. Justin From bizzaro at bc.edu Wed Sep 8 06:24:30 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] other binding project References: Message-ID: <37D6395E.32242C2@bc.edu> Justin Bradford wrote: > > > The date is early May, and I've never heard anything about this guy's > > bindings, so I imagine the project stalled. Regradless, I'm going to email > > him and see what he's up to... > > I got a response back (as well as his source). > I'm going to probably stick what he has in CVS tomorrow (assuming he > agrees to the LGPL), and then we can go from there. His code may need some > modification to conform with the Python binding proposal, but I doubt it > will be too major. So, is Michael Robinson still involved or willing to help? I take it that he gave up on it, since he just handed over all the source code. > This is a major jumpstart to this little project, as we'll be starting > with bindings that actually work (or are very near) for simple interfaces. It sure is! That's great news. Let me know when you get this in CVS, and I'll set up the permissions per our last conversation. Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From bizzaro at bc.edu Wed Sep 8 18:55:25 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] Re: biopython.org References: <37D637C0.B4EE3FE5@bc.edu> <37D6D13C.11BFCB14@bioreason.com> <37D6DD98.72A1ACEF@bc.edu> <37D6E2E3.3A001F5E@bioreason.com> Message-ID: <37D6E95D.910E0C3D@bc.edu> Andrew Dalke wrote: > > > We're making bindings to the GNU CORBA system, "ORBit". > > Out of curiosity, will it meet the OMG specs for CORBA > bindings for Python? See Yes, indeed. > > ftp://ftp.omg.org/pub/docs/orbos/99-08-02.pdf > or > http://www.informatik.hu-berlin.de/~loewis/python/pymap.htm > > I talked to Martin briefly at the last OMG meeting, and he says > that it is very close to being accepted as the OMG standard. > > In theory, if everyone (PyORBit, ILU, fnorb, etc.) met the > standard, then people would be able to pick and chose ORBs as > appropriate for their task. That's the plan :-) Thanks for the info. I'll let you know when things progress. Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From justin at ukans.edu Thu Sep 9 05:25:23 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] other binding project In-Reply-To: <37D6395E.32242C2@bc.edu> Message-ID: > So, is Michael Robinson still involved or willing to help? I take it that he > gave up on it, since he just handed over all the source code. He doesn't have much time, but he might still contribute more to this project. > It sure is! That's great news. Let me know when you get this in CVS, and I'll > set up the permissions per our last conversation. I'm going to clean up Michael's code slightly, and then put it up tomorrow (Thursday) after my lab. With that in, I say we publicize the project. With something working (or very near, which I'll check on tomorrow, as well), we'll probably get an immediate userbase and other programmers wanting to help. As for the description, I don't think it needs to be anything complex. Perhaps: pyORBit is a project to create Python bindings for ORBit. We will follow the proposed OMG bindings to insure compliance, but we will also consider non-standard extensions to improve performance, if necessary. These bindings will be distributed under the GNU LGPL license. Additionally, Michael Robinson has donated his initial implementation, which should already be functional for simple interfaces. This code is available, but should be considered a developers-only release. The structure and layout of later pyORBit releases will likely change to some degree. Regardless, bug reports and patches are welcome. http://theopenlab.uml.edu/pyorbit/ How does that sound? Perhaps another paragraph to plug The OpenLab? I figure we can send this out on Friday. That will give me time to add some links to the web page, put the code in CVS, and play with the code some. Justin From bizzaro at bc.edu Thu Sep 9 07:20:04 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] other binding project References: Message-ID: <37D797E4.12F1FAD3@bc.edu> Justin Bradford wrote: > > I'm going to clean up Michael's code slightly, and then put it up tomorrow > (Thursday) after my lab. Okay. Great. > pyORBit is a project to create Python bindings for ORBit. Does it matter to you if the letter case is pyORBit or PyORBit? We should probably have 1 sentence about ORBit, since _some_ people on _some_ of the lists might not be familiar with it. > We will follow > the proposed OMG bindings to insure compliance, but we will also > consider non-standard extensions to improve performance, if necessary. > > These bindings will be distributed under the GNU LGPL license. > > Additionally, Michael Robinson has donated his initial implementation, > which should already be functional for simple interfaces. This code is > available, but should be considered a developers-only release. Are you going to give it a release number? Does it already have one? PyORBit 0.1.0 ? > The > structure and layout of later pyORBit releases will likely change to > some degree. Regardless, bug reports and patches are welcome. > > http://theopenlab.uml.edu/pyorbit/ > > How does that sound? It sounds good to me. I think we should also mention that you are the project coordinator. > Perhaps another paragraph to plug The OpenLab? Yeah. I've got some text on hand. > I figure we can send this out on Friday. That will give me time to add > some links to the web page, put the code in CVS, and play with the code > some. Swell. It seems we're months ahead of my expectations for a CORBA interface. Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From justin at ukans.edu Fri Sep 10 04:49:19 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] other binding project In-Reply-To: <37D797E4.12F1FAD3@bc.edu> Message-ID: > > I'm going to clean up Michael's code slightly, and then put it up tomorrow > > (Thursday) after my lab. > > Okay. Great. Ok, I've fleshed out the webpage, put up a tarball (which _might_ compile for a few other people), and stuck everything in CVS. I think we're ready to go... > Does it matter to you if the letter case is pyORBit or PyORBit? PyORBit is fine. For some reason, I had thought you suggested the other way. > Are you going to give it a release number? Does it already have one? PyORBit > 0.1.0 ? I have it as 0.0.1. My plan is for 0.1.0 to be when the tree is cleaned up properly (ie. source arranged better, autoconf/automake, etc). > Swell. It seems we're months ahead of my expectations for a CORBA interface. Well, the current code is rough. It took me a couple hours to finally get enough to compile to let me run some test scripts. Then, when I tried it on another box of mine, some of the compiles bombed out... But it does work, so we're definitely off to a good start. Anyway, here's another draft of the release. Feel free to change whatever you want: PyORBit is a project to create Python bindings for ORBit, a fast, lightweight CORBA ORB used by the GNOME Desktop for application and component interaction (similar to Microsoft's COM). These bindings will be distributed under the GNU LGPL license. PyORBit will follow the proposed OMG Python bindings to insure compliance and compatibility, but we will also consider non-standard optional extensions to improve performance and add features, if necessary. Version 0.0.1 of PyORBit is already available thanks to Michael Robinson's contribution of his Python/ORBit bindings. This initial implementation should be functional for simple interfaces (assuming it compiles for you). However, this version is considered a developers-only release, as the structure and layout of later PyORBit releases will likely change to some degree. Regardless, bug reports and patches are welcome. PyORBit is coordinated by Justin Bradford . Source and a developer's mailing list can be found at: http://theopenlab.uml.edu/pyorbit/ Justin From bizzaro at bc.edu Fri Sep 10 09:06:47 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] other binding project References: Message-ID: <37D90267.475375B7@bc.edu> Okay. I'll start sending out the announcements. Cheers. Jeff Justin Bradford wrote: > > > > I'm going to clean up Michael's code slightly, and then put it up tomorrow > > > (Thursday) after my lab. > > > > Okay. Great. > > Ok, I've fleshed out the webpage, put up a tarball (which _might_ compile > for a few other people), and stuck everything in CVS. I think we're ready > to go... > > > Does it matter to you if the letter case is pyORBit or PyORBit? > > PyORBit is fine. For some reason, I had thought you suggested the other > way. > > > Are you going to give it a release number? Does it already have one? PyORBit > > 0.1.0 ? > > I have it as 0.0.1. My plan is for 0.1.0 to be when the tree is cleaned up > properly (ie. source arranged better, autoconf/automake, etc). > > > Swell. It seems we're months ahead of my expectations for a CORBA interface. > > Well, the current code is rough. It took me a couple hours to finally get > enough to compile to let me run some test scripts. Then, when I tried it > on another box of mine, some of the compiles bombed out... > > But it does work, so we're definitely off to a good start. > > Anyway, here's another draft of the release. Feel free to change whatever > you want: > > PyORBit is a project to create Python bindings for ORBit, a fast, > lightweight CORBA ORB used by the GNOME Desktop for application and > component interaction (similar to Microsoft's COM). These bindings will be > distributed under the GNU LGPL license. > > PyORBit will follow the proposed OMG Python bindings to insure compliance > and compatibility, but we will also consider non-standard optional > extensions to improve performance and add features, if necessary. > > Version 0.0.1 of PyORBit is already available thanks to Michael > Robinson's contribution of his Python/ORBit bindings. This initial > implementation should be functional for simple interfaces (assuming it > compiles for you). However, this version is considered a developers-only > release, as the structure and layout of later PyORBit releases will likely > change to some degree. Regardless, bug reports and patches are welcome. > > PyORBit is coordinated by Justin Bradford . > > Source and a developer's mailing list can be found at: > http://theopenlab.uml.edu/pyorbit/ > > Justin > > _______________________________________________ > pyorbit maillist - pyorbit@theopenlab.uml.edu > http://theopenlab.uml.edu/mailman/listinfo/pyorbit -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From bizzaro at bc.edu Fri Sep 10 10:19:24 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] announcements Message-ID: <37D9136C.9A317D0A@bc.edu> Greetings. Announcements were made to the following lists: theopenlab loci pyphy pygtk orbit comp.lang.python comp.object.corba Additionally, I posted announcements to gnome-news website freshmeat With all of these lists hit, thousands will read the announcement today. One person already subscribed: Hi Tom Jenkins! Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From TomJenkins at zentuit.com Fri Sep 10 11:00:00 1999 From: TomJenkins at zentuit.com (Tom Jenkins) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] announcements In-Reply-To: <37D9136C.9A317D0A@bc.edu> Message-ID: <199909101357.JAA00238@asprin.zentuit.com> > With all of these lists hit, thousands will read the announcement today. > > One person already subscribed: Hi Tom Jenkins! > Hi Jeff, Yeehaw! python and Orbit! -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Tom Jenkins DevIS (http://www.devis.com) And the No. 1 response that you'll need to memorize if you plan to bet your business on Windows 2000: "You want fries with that?" - Nickolas Petreley, InfoWorld, August 23, 1999 "Do not be overcome by evil, but overcome evil with good." Romans 12:21 From bizzaro at bc.edu Sat Sep 11 08:31:24 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:13 2006 Subject: [pyorbit] [Fwd: [pygtk] ANNOUNCE: Python/ORBit Bindings] Message-ID: <37DA4B9C.9FD9090@bc.edu> From justin at ukans.edu Sun Sep 12 03:23:27 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:14 2006 Subject: [pyorbit] [Fwd: [pygtk] ANNOUNCE: Python/ORBit Bindings] In-Reply-To: <37DA4B9C.9FD9090@bc.edu> Message-ID: James Henstridge wrote: >Those bindings look pretty good. It would be nice if you made them a bit >easier to configure though. After fixing the paths, the >testserver/testclient seemed to work quite well. I considered trying to work on the code a bit more before a release, but once I got it to compile and run the test scripts, I figured I might as well put something out. My plan is for 0.1.0 to be reasonably usable. I intend to have an autoconf/automake build system, clean up the tree, and try to document what things are, what works, etc. I'll also try to move it closer to the proposed spec compliance (I'm not sure how close it is, yet). I'd like to see the package distributed with gnome-python eventually, so I'm definitely be interested in any suggestions you have. >It looks like it has lots of potential. Getting it to use the interface >repository (which I hear is working in the latest versions of ORBit) and >dynamic invocation interface, and we would have a great tool for >writing scripts that use arbitrary CORBA interfaces. As this release is entirely the work of Michael Robinson, I'll be spending some time just getting up to speed with the current code. After that (version 0.1.0), I'll start working on extending the features. If there are simple and obvious improvements I can make during the restructuring (or if anyone sends patches), those will be integrated, too. Justin From james at daa.com.au Sun Sep 12 07:30:04 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:14 2006 Subject: [pyorbit] Re: pyorbit In-Reply-To: <37DA4CC3.810D90C7@bc.edu> Message-ID: I have just subscribed to the list. I don't know how much help I can be though -- I am not that experienced with the use of corba. My only background is the parts of the pdf CORBA-2.2 docs available from www.omg.org I have read (not much though -- it is almost 1000 pages long), and looking at some of the GNOME source code that uses it. As for some ideas, I think we may be able to swipe some ideas from Owen Taylor's CORBA-MICO perl package. It provides stubless interface support on both client and server ends. If we can get stubless client interfaces working, pyorbit would become a very useful as a generic scripting engine for GNOME and other CORBA based systems. More information about the perl CORBA-MICO bindings are available from: http://search.cpan.org/search?dist=CORBA-MICO I think ORBit has most of what is needed for the stubless client interfaces -- DII and the interface repository. James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ On Sat, 11 Sep 1999, J.W. Bizzaro wrote: > Hi James. > > I just got your post about PyORBit. Would you like to join the mailing list? > > http://theopenlab.uml.edu/mailman/listinfo/pyorbit > > BTW, sice you have a shell account at theopenlab.uml.edu, you have full CVS > access and can contribute to the PyORBit code. > > > Cheers. > Jeff > -- > +----------------------------+ > | J.W. Bizzaro | > | jeff@theopenlab.uml.edu | > | | > | THE OPEN LAB | > | Open Source Bioinformatics | > | | > | http://theopenlab.uml.edu/ | > +----------------------------+ > From bizzaro at bc.edu Sun Sep 12 16:52:15 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] more on corba Message-ID: <37DC127F.1939F346@bc.edu> Linux World has an introduction to CORBA in 2 parts. Here's part 1: http://www.linuxworld.com/linuxworld/lw-1999-09/lw-09-corba_1.html Of particular relevance to our interests, the demo uses the ILU Python/CORBA system. Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From james at daa.com.au Mon Sep 13 10:37:03 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyorbit changes Message-ID: I just checked in a few changes to pyorbit today. The first few were to make it compile easier if the various libraries were installed in different locations. It does this using orbit-config and libIDL-config. I also put in a number of checks to make sure we weren't creating python wrappers for nil objects. This greatly reduces the number of segfaults when playing around with the bindings. I also added a bit of code so that CORBA exceptions get converted to python exceptions. It gets round the problem in the current code where exceptions never get reset by calling CORBA_exception_init after converting the exception. Each of the routines that can cause CORBA exceptions should really be calling CORBA_exception_init themselves before any CORBA method invocations. I am also not sure whether it is worth keeping a CORBA_Environment structure associated with each object. James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ From bizzaro at bc.edu Mon Sep 13 15:35:25 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] cvsweb Message-ID: <37DD51FD.7176B8EA@bc.edu> FYI, the pyorbit module can be browsed via CVSWeb on theopenlab server: http://theopenlab.uml.edu/cgi-bin/cvsweb.cgi/pyorbit/ Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From philipd at orbitcpp.freeserve.co.uk Mon Sep 13 19:51:39 1999 From: philipd at orbitcpp.freeserve.co.uk (Phil Dawes) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL Message-ID: <37DD8E0B.AFD6AA15@orbitcpp.freeserve.co.uk> Hi Jeff et al, First of all, congratulations for getting this project started - python bindings to orbit are something I've wanted for a long time. Secondly, if it'll be any help, there's a module in gnome cvs called pyIDL which wraps libIDL for python. I wrote it because I wanted to be able to develop the idl compiler for ORBit-C++ very quickly using python, and I wasn't very keen on the design of orbit-idl (my dislike of which BTW has more to do with the prohibitive overhead of writing polymorphic OO code in C than in any design decisions made by Elliot). You can see how the library works by looking at the orbit-C++ idl-compiler source (and some sparse documentation) which you can get from: http://www.orbitcpp.freeserve.co.uk/ Do you have any plans for a dynamic client side? Something which looks up type information in the IR so that you don't have to compile IDL for client side stubs would be really cool! Anyway, enough jabbering; I'm here to help hack code if you need it. I'd be more than happy to help write the idl compiler - I've got very quick at it in the last couple of months! Cheers, Phil BTW, It's good to see James H helping on the project. Many thanks for your automake-python code - I've been using it heavily on ORBit-C++ and other python projects! From bizzaro at bc.edu Mon Sep 13 18:30:44 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL References: <37DD8E0B.AFD6AA15@orbitcpp.freeserve.co.uk> Message-ID: <37DD7B14.CA43FFC6@bc.edu> Phil Dawes wrote: > > Anyway, enough jabbering; I'm here to help hack code if you need it. I'd > be more than happy to help write the idl compiler - I've got very quick > at it in the last couple of months! I can give you a shell account on theopenlab, which will give you full access to the CVS tree. I don't know if the IDL compiler should go in a separate module or not; I'll leave it to Justin. The same offer goes to anyone who is willing to make a significant contribution to PyORBit. Contact me (jeff@theopenlab.uml.edu) about setting up an account. BTW, I don't want anyone to think that PyORBit is under my direction. I run The Open Lab and coordinate The Loci Project at the Lab, but PyORBit is Justin Bradford's baby. I'm learning CORBA myself, and being a Python freak, I hope I can learn via PyORBit. > BTW, It's good to see James H helping on the project. Many thanks for > your automake-python code - I've been using it heavily on ORBit-C++ and > other python projects! James has been a godsend for some of my projects. He may not get paid for what he does, but I hope my continuing praise of his work is some motivation ;-) Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From justin at ukans.edu Tue Sep 14 00:01:41 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL In-Reply-To: <37DD8E0B.AFD6AA15@orbitcpp.freeserve.co.uk> Message-ID: > Anyway, enough jabbering; I'm here to help hack code if you need it. I'd > be more than happy to help write the idl compiler - I've got very quick > at it in the last couple of months! The current IDL compiler is limited. Michael had started on struct support, but the source code died in a vi crash. Core dumps from that are under the misc/ directory. If you're up to writing a new compiler, then great. I had been planning to explore pyIDL when I got to the compiler, anyway. I see Jeff has given you CVS access, so feel free to make a new directory and start committing code. > Do you have any plans for a dynamic client side? Something which looks > up type information in the IR so that you don't have to compile IDL for > client side stubs would be really cool! That's our most requested feature, so we'll get it in eventually. I'm busy this week with school work, but this weekend I'm planning on a mild tree cleanup. I haven't looked at what James has committed yet for building, but I'm going to move it over to autoconf/automake. After that, I'll work on a plan for IR support and dynamic clients. Input is welcome. Also, if anyone finds limitations, bugs, or incompatibilities with the OMG spec, send them to the list. By now, there are probably people on the list who know the current code better than I do. Justin From bizzaro at bc.edu Tue Sep 14 00:41:26 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL References: Message-ID: <37DDD1F6.EB5AB776@bc.edu> Justin Bradford wrote: > > I see Jeff has given you > CVS access, so feel free to make a new directory and start committing > code. Not quite yet. Phil still needs to contact me directly, and then I'll get the account set up :-) Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From robinson at netrinsics.com Tue Sep 14 03:53:52 1999 From: robinson at netrinsics.com (Michael Robinson) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL Message-ID: <199909140753.PAA01534@netrinsics.com> I don't know if anyone's had time to take a look at it, but the current IDL compiler might seem a bit weird at first. I designed it to take full advantage of several Python features (dynamic typing, first class closures, etc.). The end result is that the IDL compiler code is very nearly WYSIWYG (once you figure out how to look at it). Basically, to add a feature to the Python binding, first, you implement a prototype binding of an IDL interface by hand, then you abstract that, and copy it over to the pyidl "macro" language. The first part is a little hard, the second part is very easy. Let me know if there are any questions or confusions. -Michael Robinson From james at daa.com.au Tue Sep 14 06:31:17 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL In-Reply-To: Message-ID: I made one change to the libIDL wrapper so that it would compile with recent libIDL's. There was a structure member that has been deprecated (they renamed it to discourage people from using it. The change may break some of pyIDL (I could not really check this because the files were a bit incoherent). On Mon, 13 Sep 1999, Justin Bradford wrote: > > Anyway, enough jabbering; I'm here to help hack code if you need it. I'd > > be more than happy to help write the idl compiler - I've got very quick > > at it in the last couple of months! > > The current IDL compiler is limited. Michael had started on struct > support, but the source code died in a vi crash. Core dumps from that are > under the misc/ directory. > > If you're up to writing a new compiler, then great. I had been planning to > explore pyIDL when I got to the compiler, anyway. I see Jeff has given you > CVS access, so feel free to make a new directory and start committing > code. > > > Do you have any plans for a dynamic client side? Something which looks > > up type information in the IR so that you don't have to compile IDL for > > client side stubs would be really cool! > > That's our most requested feature, so we'll get it in eventually. > > I'm busy this week with school work, but this weekend I'm planning on a > mild tree cleanup. I haven't looked at what James has committed yet for > building, but I'm going to move it over to autoconf/automake. I have submited the autoconf/automake macros I use in gnome-python to Tom Tromey, so they (or a modified version of them) should apear in automake-1.5. If you want to use them before then, they are available from the GNOME CVS tree in the gnome-python module. The build changes I did commit were the minimal ones to get it to use orbit-config and libIDL-config to get the compile/link flags. The other changes were a couple of NULL checks and the code to raise python exceptions when CORBA exceptions occur. This last change only required changing the checkCORBA macro. > > After that, I'll work on a plan for IR support and dynamic clients. Input > is welcome. If you are going to play around with the interface repository code, remember that you will have to pass the --ORBIfaceRepoIOR argument to the program. Something like this should be sufficient for testing: $ orbit-ird | tail /tmp/ird.ior & $ python ... >>> import CORBA >>> ird_ior = open('/tmp/ird.ior').read() >>> if ird_ior[-1] == '\n': ird_ior = ird_ior[:-1] >>> orb = CORBA.ORB_init(['foo', '--ORBIfaceRepoIOR', ird_ior], ... 'orbit-local-orb') >>> ir = orb.resolve_initial_references('InterfaceRepository') Without this, any interface repository related calls could potentially cause seg faults (because ORBit's internal pointer to the interface repository would be NULL). You can do a similar thing to test the sample orbit-name-server with the --ORBNamingIOR argument. > > Also, if anyone finds limitations, bugs, or incompatibilities with the OMG > spec, send them to the list. By now, there are probably people on the list > who know the current code better than I do. > > Justin > > > One other thing to look at is whether overloading CORBA.__dict__ as a place to register object stubs is the best way of setting things up. It works fine at the moment, but there may be better ways to register stubs. James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ From robinson at netrinsics.com Tue Sep 14 08:53:07 1999 From: robinson at netrinsics.com (Michael Robinson) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL In-Reply-To: Message-ID: <199909141253.UAA01822@netrinsics.com> >I made one change to the libIDL wrapper so that it would compile with >recent libIDL's. There was a structure member that has been deprecated >(they renamed it to discourage people from using it. The change may break >some of pyIDL (I could not really check this because the files were a bit >incoherent). What was the structure member, and in which structure? -Michael Robinson From james at daa.com.au Tue Sep 14 09:53:39 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL In-Reply-To: <199909141253.UAA01822@netrinsics.com> Message-ID: It was the f_noscript attribute of the _IDL_OP_DCL structure. The start of the structure in the IDL.h header file looks like the following: struct _IDL_OP_DCL { unsigned __f_noscript : 1; /* Deprecated */ unsigned f_oneway : 1; /* XPIDL extension (varags) */ unsigned f_varargs : 1; IDL_tree op_type_spec; IDL_tree ident; IDL_tree parameter_dcls; IDL_tree raises_expr; IDL_tree context_expr; }; James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ On Tue, 14 Sep 1999, Michael Robinson wrote: > >I made one change to the libIDL wrapper so that it would compile with > >recent libIDL's. There was a structure member that has been deprecated > >(they renamed it to discourage people from using it. The change may break > >some of pyIDL (I could not really check this because the files were a bit > >incoherent). > > What was the structure member, and in which structure? > > -Michael Robinson > > > _______________________________________________ > pyorbit maillist - pyorbit@theopenlab.uml.edu > http://theopenlab.uml.edu/mailman/listinfo/pyorbit > From robinson at netrinsics.com Tue Sep 14 10:06:58 1999 From: robinson at netrinsics.com (Michael Robinson) Date: Mon Feb 6 02:00:15 2006 Subject: [pyorbit] pyIDL In-Reply-To: Message-ID: <199909141406.WAA02531@netrinsics.com> James Henstridge writes: >It was the f_noscript attribute of the _IDL_OP_DCL structure. The start >of the structure in the IDL.h header file looks like the following: > >struct _IDL_OP_DCL { > unsigned __f_noscript : 1; /* Deprecated */ I didn't use that for anything, so presumably nothing is broken by removing it (unless weird struct member alignment issues are involved). Has anybody gotten the py-idl script to compile any sample IDL? -Michael Robinson From james at daa.com.au Tue Sep 14 10:37:38 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] pyIDL In-Reply-To: <199909141406.WAA02531@netrinsics.com> Message-ID: Actually I may have broken some stuff (I was not sure how much of the stuff in the misc directory was working, since it looked a bit corrupted). The change in libIDLmodule.c was the following: --- libIDLmodule.c 1999/09/10 07:42:31 1.1 +++ libIDLmodule.c 1999/09/13 14:08:13 1.2 @@ -292,8 +292,7 @@ IDLtree2Object(self->tree->u.idl_attr_dcl.param_type_spec), IDLtree2Object(self->tree->u.idl_attr_dcl.simple_declarations)); case IDLN_OP_DCL: - return Py_BuildValue("(iiNNNNN)", - (int) self->tree->u.idl_op_dcl.f_noscript, + return Py_BuildValue("(iNNNNN)", (int) self->tree->u.idl_op_dcl.f_oneway, IDLtree2Object(self->tree->u.idl_op_dcl.op_type_spec), IDLtree2Object(self->tree->u.idl_op_dcl.ident), It would probably cause problems for anything which was using the resulting tuple. We could put in a dummy value into the tuple, but it may be a bit cleaner to change pyIDL a bit. You would probably know best what this change would affect. James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ On Tue, 14 Sep 1999, Michael Robinson wrote: > James Henstridge writes: > >It was the f_noscript attribute of the _IDL_OP_DCL structure. The start > >of the structure in the IDL.h header file looks like the following: > > > >struct _IDL_OP_DCL { > > unsigned __f_noscript : 1; /* Deprecated */ > > I didn't use that for anything, so presumably nothing is broken by removing > it (unless weird struct member alignment issues are involved). > > Has anybody gotten the py-idl script to compile any sample IDL? > > -Michael Robinson From james at daa.com.au Tue Sep 14 22:56:25 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] Web site Message-ID: I was just looking at the pyorbit web site, and thought it would probably be a good idea to put a link to the CORBA 2.3 specification up there. It is available in postscript and pdf form at: http://www.omg.org/corba/corbaiiop.html It has good descriptions of the CORBA services and the C language mapping among other things. It seems to be written well and describes how a lot of the things we probably want to use in pyorbit work. James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ From bizzaro at bc.edu Tue Sep 14 23:19:52 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] Web site References: Message-ID: <37DF1058.D36E7572@bc.edu> Done. I posted it almost verbatim under a new 'Reference' section. Justin, you can change it if you want. We _may_ also want to post some links to online CORBA tutorials. The LinuxWorld article that covers ILU comes to mind. Cheers. Jeff James Henstridge wrote: > > I was just looking at the pyorbit web site, and thought it would probably > be a good idea to put a link to the CORBA 2.3 specification up there. It > is available in postscript and pdf form at: > http://www.omg.org/corba/corbaiiop.html > > It has good descriptions of the CORBA services and the C language mapping > among other things. It seems to be written well and describes how a lot > of the things we probably want to use in pyorbit work. > > James. > > -- > Email: james@daa.com.au > WWW: http://www.daa.com.au/~james/ > > _______________________________________________ > pyorbit maillist - pyorbit@theopenlab.uml.edu > http://theopenlab.uml.edu/mailman/listinfo/pyorbit -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From robinson at netrinsics.com Tue Sep 14 23:21:41 1999 From: robinson at netrinsics.com (Michael Robinson) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] pyIDL In-Reply-To: Message-ID: <199909150321.LAA06859@netrinsics.com> James Henstridge writes: >It would probably cause problems for anything which was using the >resulting tuple. We could put in a dummy value into the tuple, but it may >be a bit cleaner to change pyIDL a bit. You would probably know best >what this change would affect. What you need to do is decrement all the indices in the "class idl_op_dcl:" declaration in PyIDL.py. That should fix everything. -Michael Robinson From james at daa.com.au Wed Sep 15 03:01:55 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] pyIDL In-Reply-To: <199909150321.LAA06859@netrinsics.com> Message-ID: Alright, I made the change. I was just reading through the IDL -> python mapping, and it looks like a number of bits (including the DII create_request mapping) seem to require the interface repository. Does anyone know how far along it is at the moment? James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ On Wed, 15 Sep 1999, Michael Robinson wrote: > James Henstridge writes: > >It would probably cause problems for anything which was using the > >resulting tuple. We could put in a dummy value into the tuple, but it may > >be a bit cleaner to change pyIDL a bit. You would probably know best > >what this change would affect. > > What you need to do is decrement all the indices in the "class idl_op_dcl:" > declaration in PyIDL.py. That should fix everything. > > -Michael Robinson > > > _______________________________________________ > pyorbit maillist - pyorbit@theopenlab.uml.edu > http://theopenlab.uml.edu/mailman/listinfo/pyorbit > From justin at ukans.edu Wed Sep 15 03:03:51 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] pyIDL In-Reply-To: Message-ID: > I was just reading through the IDL -> python mapping, and it looks like a > number of bits (including the DII create_request mapping) seem to require > the interface repository. Does anyone know how far along it is at the > moment? Do you mean ORBit's implementation of the interface repository? I'll ask on the ORBit list. Justin From james at daa.com.au Wed Sep 15 03:35:13 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] pyIDL In-Reply-To: Message-ID: Yes I was thinking of the interface repository. From the corba docs, it was not obvious how information gets into the interface repository. I assume that there would need to be some support in the stub generator to generate code that inserts the appropriate entries when an object is created. Someone correct me if I am mistaken about this. James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ On Wed, 15 Sep 1999, Justin Bradford wrote: > > I was just reading through the IDL -> python mapping, and it looks like a > > number of bits (including the DII create_request mapping) seem to require > > the interface repository. Does anyone know how far along it is at the > > moment? > > Do you mean ORBit's implementation of the interface repository? > I'll ask on the ORBit list. > > Justin > > > > _______________________________________________ > pyorbit maillist - pyorbit@theopenlab.uml.edu > http://theopenlab.uml.edu/mailman/listinfo/pyorbit > From philipd at orbitcpp.freeserve.co.uk Wed Sep 15 08:14:03 1999 From: philipd at orbitcpp.freeserve.co.uk (Phil Dawes) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] pyIDL References: Message-ID: <37DF8D8B.C555DB82@orbitcpp.freeserve.co.uk> James Henstridge wrote: > > Yes I was thinking of the interface repository. From the corba docs, it > was not obvious how information gets into the interface repository. I > assume that there would need to be some support in the stub generator to > generate code that inserts the appropriate entries when an object is > created. > > Someone correct me if I am mistaken about this. > > James. > I think the corba interface allows you to add entries as well as query it. The interface repositories I've used all came with an 'irfeed' type program which inserts idl into the IR. I think mico has it as an option to the idl compiler. Cheers, Phil From loewis at informatik.hu-berlin.de Wed Sep 15 08:15:31 1999 From: loewis at informatik.hu-berlin.de (Martin von Loewis) Date: Mon Feb 6 02:00:16 2006 Subject: [pyorbit] pyIDL In-Reply-To: (message from James Henstridge on Wed, 15 Sep 1999 15:01:55 +0800 (WST)) References: Message-ID: <199909151215.OAA13321@pandora> > I was just reading through the IDL -> python mapping, and it looks like a > number of bits (including the DII create_request mapping) seem to require > the interface repository. If you think it shouldn't work this way (i.e. application has to provide all type information), feel free to raise an issue. OTOH, I don't think Python developers want to use the DII - especially if the ORB can create proper method signatures on-the-fly. Regards, Martin From philipd at orbitcpp.freeserve.co.uk Wed Sep 15 08:20:38 1999 From: philipd at orbitcpp.freeserve.co.uk (Phil Dawes) Date: Mon Feb 6 02:00:17 2006 Subject: [pyorbit] What's the approach? Message-ID: <37DF8F16.FEB5E302@orbitcpp.freeserve.co.uk> Hi all, I had a look at the code on the web page (and got it all to compile with some minor hacks :-) and it looks like the idl compiler generates c stubs (although I didn't get this working on my own idl - I couldn't find the right script to use). I was wondering if this was the intended overall approach? - I was thinking more along the lines of developing a dynamic marshalling interface to the orbit core and then having the idl compiler generate the skeleton code in python that talks to that (i.e. no C compiler involved). BTW, There was a discussion on the orbit list a while ago about this and if I remember right, the general conclusion was that writing a custom dynamic interface was a better idea than attempting to implement dynany in orbit, mainly because of its poor performance and difficulty of implementation. What's the plan folks? Cheers, Phil. From james at daa.com.au Wed Sep 15 08:54:28 1999 From: james at daa.com.au (James Henstridge) Date: Mon Feb 6 02:00:17 2006 Subject: [pyorbit] pyIDL In-Reply-To: <37DF8D8B.C555DB82@orbitcpp.freeserve.co.uk> Message-ID: Thanks for clearing that up. I knew of those parts of the interface, but I just wasn't sure who was responsible for calling them. I was wondering if the generated stubs were supposed to contain code for feeding the interface repository. I did not think of there being a seperate irfeed program because there isn't one in ORBit and I am a bit new to CORBA. James. -- Email: james@daa.com.au WWW: http://www.daa.com.au/~james/ On Wed, 15 Sep 1999, Phil Dawes wrote: > I think the corba interface allows you to add entries as well as query > it. > The interface repositories I've used all came with an 'irfeed' type > program which inserts idl into the IR. I think mico has it as an option > to the idl compiler. > > Cheers, > > Phil > > _______________________________________________ > pyorbit maillist - pyorbit@theopenlab.uml.edu > http://theopenlab.uml.edu/mailman/listinfo/pyorbit > From robinson at netrinsics.com Wed Sep 15 09:31:14 1999 From: robinson at netrinsics.com (Michael Robinson) Date: Mon Feb 6 02:00:17 2006 Subject: [pyorbit] What's the approach? In-Reply-To: <37DF8F16.FEB5E302@orbitcpp.freeserve.co.uk> Message-ID: <199909151331.VAA07677@netrinsics.com> Phil Dawes writes: >I had a look at the code on the web page (and got it all to compile with >some minor hacks :-) and it looks like the idl compiler generates c >stubs (although I didn't get this working on my own idl - I couldn't >find the right script to use). Yes, you're right. >I was wondering if this was the intended overall approach? It was my personal intended overall approach. I want to use Python as a middleware/second-tier scripting language. In those applications, you almost always have a well-defined interface (or, at least, should), and performance is very big consideration. I can appreciate that for other applications people would want dynamic marshalling. I think the world is big enough for both approaches, and people should work on implementing whichever approach they have the most pressing need for. -Michael Robinson From philipd at orbitcpp.freeserve.co.uk Wed Sep 15 12:30:38 1999 From: philipd at orbitcpp.freeserve.co.uk (Phil Dawes) Date: Mon Feb 6 02:00:17 2006 Subject: [pyorbit] What's the approach? References: <199909151331.VAA07677@netrinsics.com> Message-ID: <37DFC9AE.3D34956E@orbitcpp.freeserve.co.uk> Michael Robinson wrote: > > >I was wondering if this was the intended overall approach? > > It was my personal intended overall approach. I want to use Python as a > middleware/second-tier scripting language. In those applications, you > almost always have a well-defined interface (or, at least, should), and > performance is very big consideration. > Ah, right. My motivation is simply writing objects for desktop use, and maybe running python objects in the same address space as C++ objects and perl objects if there are shortcircuit performance benefits to be had. This would mean that the dynamic interface to the orbit core would be an abstraction above IIOP - maybe GIOP compliant, which the orbit core could then marshall onto the IIOP wire, or directly to other objects running in the same orb. > I can appreciate that for other applications people would want dynamic > marshalling. I think the world is big enough for both approaches, and > people should work on implementing whichever approach they have the most > pressing need for. Agreed. Cheers, Phil. From bizzaro at bc.edu Wed Sep 15 13:17:41 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:18 2006 Subject: [pyorbit] [Fwd: Interface Repository] Message-ID: <37DFD4B5.71DA188D@bc.edu> Here is a reply to Justin's post on the ORBit list. -------------- next part -------------- An embedded message was scrubbed... From: Dick Porter Subject: Re: Interface Repository Date: Wed, 15 Sep 1999 09:40:53 +0100 Size: 2549 Url: http://bioinformatics.org/pipermail/pyorbit/attachments/19990915/069a1724/attachment.mht From bizzaro at bc.edu Wed Sep 15 22:44:38 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:18 2006 Subject: [pyorbit] server was down Message-ID: <37E05996.28F21CB3@bc.edu> Greetings. You may not have noticed, but theopenlab server was down for a few hours this afternoon. Certain upgrades that had to be made broke a number of services, including network access, so I upgraded the whole OS to RedHat 6. Everything seemed to be functioning until I tried the mailing lists. It turns out we upgraded to the version of sendmail that uses a 'restricted shell', and it took a few more hours to figure that out. Let me know if you find anything else that doesn't work. Sorry for any inconvenience. Cheers. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From bizzaro at bc.edu Sun Sep 19 20:43:33 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:18 2006 Subject: [pyorbit] ORBit module for Perl released Message-ID: <37E58334.614D7AB2@bc.edu> This comes from the Gnotices Web page: Posted by Owen Taylor on Sunday September 19, @08:11PM I've made an initial release of dynamic Perl bindings for ORBit available. They are available here. ftp://ftp.gnome.org/pub/otaylor/ Though far from complete, they pretty useable already. here is also a GNOME::GNORBA module there for using CORBA::ORBit within GNOME. The nifty thing about CORBA::ORBit is that you don't need to to create the stubs/skeletons as a separate step. You can either point it directly at an IDL file that it will parse directly, or it can it can get the information from an interface repository. (Not the ORBit IR, yet.) It is also very compact in terms of the amount of code necessary to implement clients and servers. CORBA::ORBit will work out of the box with the Gtk:: Perl module, and an example of this is included. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From bizzaro at bc.edu Tue Sep 21 22:38:29 1999 From: bizzaro at bc.edu (J.W. Bizzaro) Date: Mon Feb 6 02:00:19 2006 Subject: [pyorbit] server's back Message-ID: <37E84125.8F82FC01@bc.edu> Greetings. We had a small glitch, but everything is back up and running. My apologies. Jeff -- +----------------------------+ | J.W. Bizzaro | | jeff@theopenlab.uml.edu | | | | THE OPEN LAB | | Open Source Bioinformatics | | | | http://theopenlab.uml.edu/ | +----------------------------+ From justin at ukans.edu Wed Sep 29 12:34:22 1999 From: justin at ukans.edu (Justin Bradford) Date: Mon Feb 6 02:00:19 2006 Subject: [pyorbit] Re: PyOrnit In-Reply-To: <199909291337.PAA02136@linoleum.inria.fr> Message-ID: On Wed, 29 Sep 1999, Bertrand Guiheneuf wrote: > there is no sign of developpement on the ml archives. > What's up with the ORbit bindings ? That's because I overestimated my free-time last week. I haven't had the chance to do some things I had planned. So development (on my part, at least) is temporarily stalled. I'll try to put in some time regularly from now on. > I am considering writing the interface of Evolution (Gnome Mailer) > in ORBit but this is highly conditioned by the availability of > an ORBit binding. Well, have you tried using the current PyORBit code? I can try to address specific limitations you find. > btw, is there any cvs server where to get PyORBit ? Yes, but I don't think anonymous CVS has been setup yet. I'll check on that status. Also, I'll try to release another snapshot soon. Justin